Agero Machine Learning Engineer Interview Questions & Guide 2026

The 60-second brief

At Agero, the Machine Learning Engineer role—often positioned as a Principal Optimization and Machine Learning Engineer—is a pivotal technical position that sits at the intersection of data science, operations research, and large-scale software engineering. You will be joining a team responsible for the next-generation Dispatch System, a critical platform powered by Swoop that manages millions of roadside events annually. This system determines which service provider gets which job, when, and why, directly impacting the safety of stranded drivers and the efficiency of the network.

For a design discussion, a catalogue of architectures is worth less than the ability to turn a vague requirement into a data model and an API contract. A box diagram with no schema under it collapses at the first follow-up question.

Focus areas

• Trace a symptom to a mechanism under load
• Choose indexes from the query's access path
• Make every write idempotent under retry

Overview

This role goes beyond standard predictive modeling. You are expected to fuse short-term and long-term horizon optimizers to solve complex logistical problems. Your work will involve architecting end-to-end Python services, building prediction models using techniques like gradient boosting and deep learning, and integrating them into constrained optimization frameworks. You will be instrumental in rethinking the vehicle ownership experience by transforming manual dispatch processes into digital, transparent, and connected solutions.

For Agero, this position is strategic. You are not just optimizing a metric in a vacuum; you are balancing cost efficiency against service-level agreements (SLAs) and Net Promoter Scores (NPS). The solutions you build will be deployed on AWS and must operate in real-time, handling the complexity of a massive B2B white-label network that serves over 150 million vehicle coverage points.

The Agero interview process

Agero candidates report 5 rounds · ≈ 4-6 weeks. The stages below are what candidates describe, not a published process.

Recruiter Screen

Before anything technical happens, someone has to decide which rung of the ladder your loop is calibrated to, and that decision sets the bar for every round after it. It comes from how you describe scope, not from your title, because titles do not convert cleanly between companies. The weak version of the answer is team size and years. The strong version names the largest change you shipped where nobody reviewed the design, what would have broken if you had been wrong, and what you were paged for. Get the level said out loud on this call, because the range and the loop both follow from it.

What to demonstrate

• Whether the scope in your own account maps onto a level the team actually has an opening at, so a mismatch ends the process cheaply rather than after four interviewers have spent a day
• Whether your title needs re-mapping: the same word describes very different amounts of independent decision-making at a twenty-person company and a ten-thousand-person one
• Whether your compensation expectation can be filled at that level in the structure the role pays in, which is why the number gets asked for before any engineer is scheduled

How to prepare

• Write down two changes from the last two years: the largest one you designed with nobody reviewing the design, and the largest one where someone more senior did. Lead with the first when scope comes up, and be ready to say which parts of the second were yours
• Ask which level the loop is calibrated to and what changes at the level above it, then plan your weeks from that answer rather than from the posting
• Settle a total-compensation range beforehand with the split named, base against bonus against equity and its vesting period, so a question about numbers gets a number instead of the word market

Technical Screen

Input bounds are the part of the prompt most often skimmed, and they usually contain the answer. They tell you which complexity class is admissible, which narrows the search before you have thought about the problem itself. As a rough planning figure, a compiled language does on the order of 10^8 simple operations per second and an interpreted one roughly an order of magnitude less. So n up to about twenty admits enumerating subsets, a few thousand admits a quadratic pass, and a million admits neither: you need near-linear, or linear with a log factor. If the bounds are missing, ask for them.

What to demonstrate

• Whether the approach is justified by the stated input size rather than by whichever pattern you recognised first
• Whether you ask about the properties that change the algorithm: whether the input arrives sorted, whether duplicates occur, whether values are bounded integers, whether it all fits in memory
• Whether you can name the bottleneck in your own solution and what would remove it, even when you deliberately leave it in place
• Whether a claimed speedup is real, since memoising a recursion only helps when subproblems genuinely overlap and the state can be keyed cheaply

How to prepare

• For each algorithm you rely on, write down the largest n it handles in roughly a second, then check two of those figures by timing them in the language you will actually type in
• For two weeks, write one line naming your target complexity and the bound that justifies it before you write any code, then compare that line with what you ended up submitting
• Practise the conversion backwards: given a required O(n log n), list the mechanisms that get you there (sorting, a heap, an ordered map, divide and conquer) and choose by what the problem needs to query, not by what you used last

Comprehensive Loop

Where the day includes a partner from product, design or data, that conversation is weighted like the technical ones and prepared for least. They are deciding one thing: whether having you in the room makes their decisions cheaper. That means options with costs attached, not implementation detail and not "it depends". An estimate someone can plan against — a range, the assumption that would push it to the high end, and what you would drop to hit the low one — is worth more than a confident single number, which everyone present already knows is wrong.

What to demonstrate

• Whether an estimate comes as a range with the assumption most likely to break it, and states what a specific scope cut would actually buy
• Whether a technical constraint is handed over as a choice with consequences on their side, rather than as a verdict they have no standing to argue with
• Whether you establish what decision is on the table before proposing anything
• Whether risk is raised while it can still change the plan, with the trigger that would confirm it, instead of reported afterwards as a slip

How to prepare

• Take a project that shipped late and write the two-sentence warning you could have given three weeks earlier, naming what you would have needed decided at that point
• Rehearse one estimate out loud until it arrives in three parts: the range, the single assumption that would blow it, and the smallest thing you would cut to protect the date
• Rewrite an objection you have actually made — the "we can't do that" version — as two options with their costs, so the choice ends up with the person who owns it

System Design Interview

Because the format is not fixed, the first job in the room is classification. Listen to the opening question and decide what it is: a probe into work you have already described, a fresh problem to solve now, or a conversation about how you operate. Each wants a different register, and the common failure is forcing a rehearsed structure onto a question that did not ask for it. Running a full design ritual on a ten-minute debugging question reads as not listening. When you cannot tell which it is, ask how long they want to spend and answer at that depth.

What to demonstrate

• Whether the shape of your answer matches the question, so a yes-or-no gets answered before it is justified and an open prompt gets a direction before a detour
• Whether you check how much depth is wanted instead of deciding for them, and whether you stop when the answer is complete rather than continuing until someone interrupts
• Whether you can be redirected in the middle of an answer without restarting it from the beginning
• Whether a question outside your experience gets an honest boundary followed by reasoning from what you do know, instead of a confident answer with nothing behind it

How to prepare

• Rehearse one project at three lengths, roughly thirty seconds, three minutes, and a full walkthrough at the depth of a design review, and practise switching between them when someone interrupts mid-telling
• Have someone ask you five questions of deliberately mixed type in one sitting without telling you the types, and score only whether you identified each one correctly before you started answering
• Draft the sentence you will use to check depth, along the lines of asking whether the short version is useful here or they want the detail, and use it in a real conversation this week so the day of the round is not its first outing

Behavioral Assessment

This round is deciding whether a change you make without supervision can be allowed to reach production. It is scored on what you knew at the moment you decided, not on how it turned out, so a story that opens with the result and works backwards reads as luck retold as judgement. Say what the options were, what you did not know, what you did to shrink the unknown before committing, and what you accepted as the worst plausible case. The detail that separates answers is a bound: how many users, how much data, and for how long, if you had been wrong.

What to demonstrate

• Whether the reasoning you give was available at the time you decided rather than after the result came in, since a story whose deciding evidence arrived later describes an outcome and not a judgement
• Whether you can put units on the exposure (users, rows, minutes of degraded service) and whether the containment you chose actually bounded it: a canary bounds the request path it fronts, while a background job writing to a shared table reaches every user regardless of which version served their requests
• Whether the reversal path existed before you shipped or was improvised during the incident, and whether it restores state or only stops further damage

How to prepare

• For your three largest changes, write down the one thing you would have had to be wrong about for it to fail, and what your best estimate of it was on the day you shipped. If you never held an estimate, that is the gap the follow-up questions will find
• Write the undo procedure for one of those changes as it existed at the time, then mark which steps restore data and which only stop new damage. Turning a flag off or reverting a deploy ends the new writes; rows already written come back only from a copy you kept, and a dropped column comes back empty unless something outside the schema holds the values
• Rehearse one story from the decision point forward and stop before the outcome, then have someone ask what you would do next. If the story only works with the ending attached, it is an anecdote rather than a decision you can defend

Where candidates lose points

Choosing an index from the columns a query mentions rather than from how it filters and orders

A composite B-tree index on (a, b, c) can be seeked only as a left prefix: equality on a, then equality on b, then a range or an ordering on c. A query that filters on b alone cannot seek into it at all and at best gets a full scan of the index; a query that filters a and ranges on b gets no benefit from c, because the index is only sorted by c within a fixed (a, b) pair. The practical consequence is that one index per column is close to useless for multi-predicate queries while a single correctly ordered composite index turns a scan into a lookup. The ordering half is what gets missed: if the index cannot satisfy the ORDER BY, the database must read every matching row and sort before the limit can apply, so a LIMIT 20 over a million matching rows still reads a million rows.

Assuming an isolation level prevents the anomaly you actually have

Isolation levels are named by the SQL standard but implemented differently, so any claim about one is only true of a named engine. PostgreSQL defaults to READ COMMITTED, where every statement takes a fresh snapshot, so two statements inside one transaction can legitimately disagree about the same row. Its REPEATABLE READ is snapshot isolation: it removes non-repeatable and phantom reads but permits write skew, where two transactions each read a set, each conclude their own write is safe, both commit, and the combined result violates a constraint that no single row expresses. Only SERIALIZABLE closes that, and it closes it by aborting a transaction with a serialization failure (SQLSTATE 40001), which means the guarantee is theoretical unless the application has a retry loop. InnoDB's REPEATABLE READ is a different mechanism again - plain SELECTs read a consistent snapshot while locking reads and writes see the latest committed row - so a read-modify-write inside one transaction can act on a value that the transaction's own earlier SELECT never returned.

Hardcoding to the sample inputs

Solve the stated problem rather than the two examples; special-casing a literal to make a sample pass is obvious immediately and reads as either a misunderstanding or an attempt to fake progress. If you genuinely cannot generalise yet, say which part is a stub and what would replace it.

Starting work without saying what you are about to spend time on

State the plan before executing it: the approach, roughly how long it will take, and what you intend to leave hand-waved. That gives the interviewer a chance to redirect you in ten seconds rather than watching you spend fifteen minutes on the wrong sub-problem.

Questions & practice

"Describe a situation where you identified a failure mode in a production model. How did you fix it and prevent it from recurring?"

Medium · machine learning fundamentals

Approach

1. State the learning problem: the label, the unit of prediction and how the model is used.
2. Say how you would validate it, and where leakage could enter the split.
3. Pick the metric from the cost of each error type, not from habit.

Follow-up

• How would you know the model is overfitting?
• What changes if the classes are heavily imbalanced?

"Design a system that predicts the probability of a provider accepting a job. How do you integrate this probability into the dispatch decision?"

Medium · machine learning fundamentals

Approach

1. Name the simplest model that could work and what would make you move past it.
2. Pick the metric from the cost of each error type, not from habit.
3. State the learning problem: the label, the unit of prediction and how the model is used.

Follow-up

• Where could label leakage enter this setup?
• What changes if the classes are heavily imbalanced?

"How would you model the trade-off between assigning a closer (more expensive) provider versus a farther (cheaper) one?"

Medium · machine learning fundamentals

Approach

1. Pick the metric from the cost of each error type, not from habit.
2. Name the simplest model that could work and what would make you move past it.
3. Say how you would validate it, and where leakage could enter the split.

Follow-up

• What changes if the classes are heavily imbalanced?
• Where could label leakage enter this setup?

"Explain how you would use OR-Tools to solve a constraint satisfaction problem."

Medium · coding and algorithms

Approach

1. Choose the data structure from the access pattern, not from familiarity.
2. State the target complexity and say which constraint rules the naive version out.
3. Walk one small example through your approach before writing the whole thing.

Follow-up

• What is the worst case, and how likely is it on real data?
• Which test case would catch an off-by-one here?

"If our optimization solver takes too long to run in real-time, what heuristics or approximations would you use to speed it up?"

Medium · coding and algorithms

Approach

1. Name the brute-force solution and its complexity before improving on it.
2. Choose the data structure from the access pattern, not from familiarity.
3. Walk one small example through your approach before writing the whole thing.

Follow-up

• Which test case would catch an off-by-one here?
• How does this change if the input no longer fits in memory?

Canonicalise a request body into a stable idempotency fingerprint

Medium · parsing · canonicalisation · hashing

idempotency_key.request_fingerprint is a SHA-256 over the method, path and canonicalised body, and a retry whose fingerprint differs must be rejected with 422 rather than served the stored response. Write the canonicaliser. Bodies are JSON up to 256 KB nested at most 32 levels; clients vary key order, whitespace and unicode escaping, and some send 64-bit ids as JSON numbers. Produce a deterministic byte string such that semantically identical bodies match and any semantic difference does not. State your complexity and name two normalisations you refuse to perform.

Approach

1. Parse once into a tree, then re-serialise under fixed rules: object keys sorted, array order preserved, one escaping convention, no insignificant whitespace. Parsing is O(n) and sorting keys is O(k log k) per object, so O(n log n) overall with O(depth) stack, and the 32-level cap is enforced during parsing because hostile nesting is how a canonicaliser becomes a stack overflow.
2. Sort keys by their UTF-8 bytes and say why the obvious implementation is wrong in some runtimes: a default string comparison that orders by UTF-16 code units places surrogate pairs, meaning code points from U+10000 up, below U+E000 to U+FFFF, which is not UTF-8 byte order, so two services written in different languages disagree on the same document.
3. Do not re-encode numbers through a double. IEEE-754 binary64 represents integers exactly only up to 2^53, so normalising a 19-digit id through a float changes it, and 1 against 1.0 cannot be reconciled without deciding whether they are the same value. Preserve the literal token, and require ids as strings at the API boundary if you want them comparable.
4. Reject duplicate keys rather than picking one. JSON permits them and parsers disagree, most keeping the last, so any choice you make ties the fingerprint to a parser detail that the code handling the request does not necessarily share.
5. Frame the hash preimage so concatenation cannot collide: delimit or length-prefix the method, path and body, otherwise one request's fields can be rearranged into another request with the same byte stream and the same fingerprint.
6. Name the refusals and their consequence: no case folding, no dropping of null-valued keys, no Unicode normalisation. Each makes two different requests fingerprint alike, and the resulting failure is the worst one this table has, since the second request is answered with the first one's stored response and its effect never happens.

Worked solution (25 min)

1. Write the serialiser: recursive emit with a depth counter, objects sorted by UTF-8 key bytes, arrays in order, strings escaped by one fixed rule, numbers emitted as their original token.
2. Run it over three bodies: the same object with keys reordered, the same object with \u0041 written as A, and one with a nested array reversed. The first two must produce identical bytes and the third must not.
3. Take the id 9007199254740993, round-trip it through a double, show it returns as 9007199254740992, then state the rule that prevents this.
4. Define the hash preimage explicitly with its delimiters, and construct a pair of (path, body) inputs that would collide without them.

Expected result: A canonicaliser that is O(n log n) in body size with an enforced depth cap, sorts keys by UTF-8 byte order, preserves array order, keeps number literals verbatim, rejects duplicate keys, and feeds a delimited preimage to SHA-256, together with a stated list of normalisations deliberately not performed and the failure each would cause.

Verify it yourself

• Key reordering and whitespace changes produce identical fingerprints while reordering an array does not.
• A body containing an integer above 2^53 fingerprints identically after a parse and re-serialise round trip.
• Two requests differing only in a field the endpoint ignores produce different fingerprints, and you can say why that is the safe direction to err.

Follow-up

• A client sends the same logical request with an extra field your API ignores. Same key, different fingerprint, so you return 422. Is that the right answer?
• Where does the fingerprint get computed relative to request decompression and the body-size limit?
• The endpoint takes 1,000 requests per second with 256 KB bodies. What does hashing cost, and does it belong at the edge or in the core service?

Write the update path that detects a concurrent edit

Medium · optimistic concurrency · transactional outbox · audit history

resource carries version INT NOT NULL DEFAULT 1. resource_revision holds revision_id, resource_id, version, actor_user_id, change_kind, patch JSONB, request_id, created_at with UNIQUE (resource_id, version). outbox_event holds aggregate_type, aggregate_id, aggregate_version, event_type, payload, status. A PUT carries the version the client read. Write the exact statements for the single transaction that applies the edit, records the revision and enqueues 'resource.updated', and give the handler's branch on zero affected rows. Then say what PostgreSQL 16 does under READ COMMITTED when two of these updates hit one row at once.

Approach

1. One transaction, three writes, no network call inside it: UPDATE resource SET title = $3, version = version + 1, updated_at = now() WHERE resource_id = $1 AND tenant_id = $4 AND version = $2; then INSERT the resource_revision row at version $2 + 1; then INSERT the outbox_event row at the same aggregate_version. The event goes to a table rather than a broker because no transaction spans both.
2. Branch on the affected-row count before doing anything else. Zero has three causes — stale version, wrong tenant, row gone — so re-read once and map to 409 carrying the current version, or 404 for an id outside the caller's tenant, which also stops the endpoint confirming that another tenant's id exists.
3. State the engine behaviour instead of assuming it. Under READ COMMITTED the second UPDATE blocks on the row lock, and when the first commits PostgreSQL re-evaluates the WHERE clause against the newly committed row, so the version predicate now fails and the statement reports zero rows. Under REPEATABLE READ the identical collision raises SQLSTATE 40001 instead, so the handler must fold both shapes into one conflict response.
4. Keep UNIQUE (resource_id, version) even though the predicate already serialises writers. It is what makes a lost update unwritable if any other path ever reaches the revision table, and it converts a logic bug into 23505 rather than into a silently missing history row.
5. Refuse to auto-retry the whole PUT. A retry re-reads the winner's state and reapplies an intent formed against data that no longer exists — the silent overwrite the version token was added to detect. Return the conflict; merge field-wise only if the patches are provably disjoint.
6. Note that now() is the transaction timestamp in PostgreSQL, so resource.updated_at, the revision's created_at and the outbox row share one instant, which is what later makes reconciliation between the three tables unambiguous.

Follow-up

• A client sends the version it read ten minutes ago and the resource has moved three versions. What is in your 409 so it can resolve the conflict without a full re-fetch?
• Two editors, two disjoint fields, no overlap. Does your answer still refuse the second write, and should it?
• Every write now touches a second hot table. How do you keep the outbox insert and its partial index from becoming the write bottleneck at 1.2k writes/second?

Explain why the owner filter ignores the listing index

Medium · composite index · query plans · sort elimination

The only index on resource is (tenant_id, status, updated_at DESC, resource_id DESC). A new endpoint returns one user's resources across all statuses, newest created first: WHERE tenant_id = $1 AND owner_user_id = $2 ORDER BY created_at DESC LIMIT 20. On a tenant with 2M rows it takes 900 ms and EXPLAIN shows a sort above a large scan. Explain precisely why the existing index cannot serve it, give the index that can, and state which of these the new index still will not help: owner_user_id alone across tenants; the same query ordered by updated_at. PostgreSQL 16.

Approach

1. Separate the two jobs an index does. For filtering, a composite btree is seekable only on a left prefix, so with no predicate on status the scan can at best range over tenant_id and test owner_user_id per row; PostgreSQL 16 has no btree skip scan to jump the unconstrained column.
2. For ordering, the index is sorted by (status, updated_at) within a tenant and not by created_at, so the LIMIT cannot stop early: every matching row is read and then sorted. That is the 'Sort Method: top-N heapsort' line, and it is why the plan reads 2M rows to answer with 20.
3. Derive the replacement from the access path — equality, equality, then the ordering column: CREATE INDEX CONCURRENTLY ON resource (tenant_id, owner_user_id, created_at DESC). The scan seeks to the (tenant, owner) range and walks 20 entries in order, so the Sort node disappears along with the row-read.
4. Treat INCLUDE (title, status) as conditional, not free. An index-only scan still visits the heap for any row whose page is not marked all-visible, so on a table taking 1.2k writes/second the win depends on autovacuum keeping the visibility map current, and the wider index costs more on every insert.
5. Answer the two negatives explicitly. owner_user_id alone is not a left prefix of the new index, so it degrades to a full scan of the index at best. Ordered by updated_at, the query still seeks on the (tenant, owner) pair but must sort, because only created_at is ordered within that pair.
6. Measure both sides with EXPLAIN (ANALYZE, BUFFERS) and compare estimated against actual rows at the lowest node — a 2M-versus-200 misestimate there is usually what chose the plan, and adding an index will not fix a statistics problem.

Worked solution (25 min)

1. Load 2M resource rows across 5k owners in one tenant, run the query under EXPLAIN (ANALYZE, BUFFERS), and record the node reading the most rows plus the Sort Method line.
2. Create (tenant_id, owner_user_id, created_at DESC) concurrently and re-run, confirming the Sort node is gone and actual rows fall to about 20.
3. Run the two negative cases and capture the plan for each.
4. Re-run the original tenant listing query to confirm the new index has not displaced the index that query depends on.

Expected result: With (tenant_id, owner_user_id, created_at DESC) the plan is an Index Scan with no Sort, roughly 20 rows and a handful of buffers, sub-millisecond. owner_user_id alone still scans; ORDER BY updated_at seeks on the pair and then sorts that owner's rows.

Verify it yourself

• Actual rows at the leaf node is on the order of the LIMIT, not the tenant's row count.
• No Sort node appears above the index scan.
• Buffers read stays constant as one owner's row count grows; if it grows with it, the ORDER BY is not being satisfied by the index.
• pg_stat_user_indexes shows idx_scan climbing on the new index and not on the listing index for this endpoint.

Follow-up

• 90% of rows are status='active'. Would a partial index WHERE status = 'active' change your answer, and for which of the three queries?
• A dashboard runs this for 40 owners in one page load. What changes about the design?
• How do you roll this index out on a table taking 1.2k writes/second, and what does it cost on every insert from then on?

"We want to A/B test a new dispatch algorithm. How do you design the experiment to ensure we don't negatively impact stranded drivers in the control group?"

Medium · ml system design and production

Approach

1. Separate the offline training path from the online serving path.
2. Fix the product goal and the online metric before choosing any model.
3. Say where features come from at serving time and how they match training.

Follow-up

• What happens when a feature is missing at serving time?
• How would you roll the new model out safely?

"How would you architect a feature store for real-time vehicle location data?"

Medium · ml system design and production

Approach

1. Fix the product goal and the online metric before choosing any model.
2. Name what you would monitor after launch and what triggers a retrain.
3. Separate the offline training path from the online serving path.

Follow-up

• How would you roll the new model out safely?
• What happens when a feature is missing at serving time?

Specify the outbound webhook contract a customer endpoint can verify

Medium · webhooks · hmac · at-least-once · ordering

Egress delivery sends resource.created, resource.updated and resource.archived to roughly 40k customer endpoints at about 1.5k deliveries/second, with a per-destination concurrency cap of 4 and a 10 second connect-plus-read timeout, at-least-once from the relay. The receiver is code you do not control, often behind a proxy that may re-encode JSON. Specify the payload envelope, how the receiver verifies authenticity and rejects replays, what ordering it may assume, how it deduplicates, which status codes it should return and what each means to us, and the retry, circuit and replay behaviour.

Approach

1. Send the fact, not a pointer: event_id, event_type, aggregate_type, aggregate_id, aggregate_version, occurred_at and the payload copied from outbox_event. A receiver told only that resource 42 changed has to call back, and the call back returns a newer state than the event describes - that is precisely how a consumer applies changes out of order while believing it is following them.
2. Sign the exact bytes on the wire: HMAC-SHA256 over the timestamp, a separator, and the raw body, in a header, compared in constant time before the body is parsed. Instruct receivers to verify the raw bytes, never a re-serialised copy, since a proxy that reorders keys or changes whitespace invalidates a signature computed over re-encoded JSON. Reject a timestamp outside a tolerance - five minutes is typical - and keep a short seen-set of event_ids, because a capture replayed inside the tolerance still verifies.
3. Support key rotation by sending several signature values in one header during the overlap window and publishing that any one that verifies is sufficient. Without that rule a key rotation is a synchronised outage across every destination at once.
4. State the guarantee as it actually is: at-least-once and unordered, including within a single aggregate_id. The outbox is ordered per aggregate at the source, but the relay holds up to 4 requests in flight per destination and backs off each retry independently, so a version 6 delivery that needs one retry lands after a version 7 that succeeded on its first attempt. Publishing per-aggregate ordering we do not enforce is worse than publishing none, because a receiver that trusts it overwrites new state with old. Receivers deduplicate on event_id and apply conditionally - UPDATE ... WHERE aggregate_version < :incoming, or the same comparison under a row lock on the aggregate - which discards stale redeliveries and stays correct when two deliveries for one aggregate are processed at the same time. A duplicate and an out-of-order delivery then have the same outcome: no-op.
5. Define the response contract in the receiver's terms. 2xx means durably accepted, and accepting means writing it down - a receiver that does its processing inline inside our 10 second timeout will be retried while it is still working. 4xx is permanent and we stop. 429, 5xx and timeouts are retried with capped exponential backoff and full jitter within the per-destination concurrency cap, and a circuit opens after a run of consecutive failures and half-opens on a schedule, so a permanently dead endpoint costs one probe per interval instead of a growing queue.
6. Give them a recovery path that does not involve us holding a queue for everyone: a visible parked state, and a replay endpoint keyed by event_id or time range, so a receiver that was down for two hours asks for the gap on its own schedule.

Worked solution (30 min)

1. Write the envelope fields and mark which are needed for deduplication and which for discarding a stale delivery.
2. Write the signature string construction and the receiver's verification steps in order, including the constant-time compare and the tolerance check.
3. Write the receiver's idempotent apply: dedupe on event_id, then the version-conditional write that discards stale and concurrently delivered redeliveries.
4. Write the status-to-action table from our side: which codes retry, which stop, and what opens the circuit.
5. Describe the parked state and the replay call, including what the customer must supply.

Expected result: A self-contained signed envelope, verification over raw bytes with a replay window and a seen-set, an at-least-once guarantee that states delivery is unordered even within one aggregate, a version-conditional apply, a status table that distinguishes permanent from transient, and a customer-driven replay path.

Verify it yourself

• Delivering the same event twice leaves the receiver's state identical to delivering it once.
• Delivering version 7 and then version 6 for one aggregate leaves version 7 applied.
• A destination that always times out costs one probe per circuit interval rather than a retry queue.

Follow-up

• A customer insists an event never arrived. What do you check, in order, and which table settles the argument?
• A receiver cannot store a version and asks us to deliver their events in order. What would the relay have to give up to offer that, and what does one failing event then do to the rest of their stream?

Edge instances grow 400 MB per hour until the nightly restart

Hard · memory leak · heap analysis · metric cardinality · gc

Edge API instances start at 700 MB resident and grow about 400 MB/hour; a nightly rolling restart has hidden it for weeks. Growth continues unchanged when request rate halves overnight, p99 degrades in the last hours before an instance is recycled, and heap used immediately after a forced full GC rises monotonically. The service holds no product state. Name the discriminating measurement that separates the plausible causes, give the most likely cause, and give the fix and how you would verify it.

Approach

1. Separate resident memory from live heap first, because they fail differently. Resident size can grow from fragmentation, native buffers or thread stacks while the heap is flat; heap used after a full GC rising monotonically is the measurement that says objects are reachable and not being released. You already have it, so this is retention, not fragmentation, and that closes off half the candidate list.
2. Use the rate's independence from traffic as the discriminator. Growth that continues at half the request rate rules out per-request objects that are merely slow to collect and points at a structure that grows with distinct values observed rather than with call volume. Write the candidates that have that property: a metrics registry keyed on a high-cardinality label, an unevicted cache, an interner, a per-key lock map.
3. Take two heap snapshots an hour apart and diff by retained size, reading the dominator tree, not by allocation count or instance count. Expect one root holding a map with millions of entries, then follow the reference chain to the code that inserts and never removes. Allocation profilers point at churn, which is the wrong signal here.
4. The candidate that fits this service is an observability label carrying an identifier, such as a request path recorded before templating so that /v1/resources/48213 becomes its own metric series. That grows with distinct ids seen, is independent of rate, and explains the late p99 degradation, since GC cost rises with the size of the live set.
5. Fix by bounding cardinality at the source: template the path to /v1/resources/{id} before it becomes a label, move tenant id from a label to a log field or an exemplar, and cap the registry with a bounded map that evicts. Add a cardinality ceiling that fails loudly in a lower environment rather than growing quietly in production.
6. Verify with a soak rather than a restart. Hold one instance out of the nightly recycle for 48 hours with the fix and compare post-GC heap and series count against an unfixed control taking the same traffic.

Follow-up

• Post-GC heap is now flat but resident size still creeps. What are you looking at, and does it matter?
• How would you have detected this before an OOM, given the nightly restart masked the trend?
• That label is what makes one dashboard useful. How do you keep the dashboard and lose the leak?

Suggested practice checklist

Roughly ninety minutes on weeknights with one longer weekend block. The plan cuts scope rather than compressing everything, on the assumption that one thing finished per night beats four half-started.

Day 1 — Fix the scope and take a cold baseline

• Read the role description and write the three things the loop will almost certainly test, then write an explicit not-doing list and keep it visible all week.
• Take one twenty-five-minute coding problem and one fifteen-minute design prompt cold, and write the single sentence naming what blocked each, because those two sentences decide where the remaining evenings go.
• Set the week's rule: one thing finished every night, including the night you only have forty minutes.

Deliverable: A one-page scope with a not-doing list and two cold attempts, each carrying one sentence on what blocked it.

Day 2 — One pattern, written three times from blank

• Choose the single pattern most likely to appear in your loop and write it three times from an empty file rather than editing the previous attempt.
• On the third pass, write the invariant as a comment before the loop body and the complexity before the first line of code.
• Stop at ninety minutes even if the third version is imperfect, and write the one thing you would fix given another hour.

Deliverable: Three independent implementations of the same pattern plus a note on what changed between them.

Day 3 — One design, only to the depth you can defend

• Take one system shape and go only as far as requirements, interface and data model, refusing to draw a box you could not survive a follow-up about.
• Attach one number to each non-functional requirement, deriving it rather than asserting it, and write the assumption the number rests on.
• Write the one tradeoff you are choosing against and the observation that would make you reverse it.

Deliverable: One design at interface-and-schema depth with derived numbers and one written reversible tradeoff.

Day 4 — Only the fundamentals you will have to defend

• Write, in under two hundred words each, the answers to the two questions that follow almost any implementation: why this structure and not the obvious alternative, and what happens to this code at a hundred times the input.
• Write what an index actually costs: faster lookups on the indexed columns against a write that now maintains a second structure, plus the cases where the planner declines to use it anyway, low selectivity, or a predicate wrapping the column in a function.
• Delete any answer you cannot deliver aloud in under a minute, since an answer that needs reading is not an answer you have.

Deliverable: Three written answers, each under two hundred words and each timed aloud.

Day 5 — Your own work, timed

• Write a ninety-second and a four-minute version of your main project and time both aloud rather than reading them.
• Prepare the two follow-ups that always come: what you would do differently, and how you knew it worked.
• Put one number in the first sentence and be ready to say exactly where it came from and what it excludes.

Deliverable: Two timed narratives with one defensible number in the opening line.

Day 6 — The one full rehearsal, in the weekend block

• Run a sixty-minute mock covering a coding round and a design round in one sitting with no break, because sustained attention is the thing evenings have not trained.
• Immediately afterwards, and before hearing any feedback, write the three moments you lost the thread.
• Spend the rest of the block only on those three moments, and on nothing you merely feel shaky about.

Deliverable: Mock notes naming three failure moments with a specific fix written under each.

Day 7 — Taper

• Write the twenty-minute warm-up you will actually do on the morning: one problem you can already solve from a blank file, one design you can narrate, and nothing you have never seen.
• Re-read only your own notes from this week and open no new material.
• Write the logistics down: the editor or shared document you will be working in, whether execution and lookups are permitted, and the sentence you will use when you do not know something.

Deliverable: A one-page card holding the design structure, the project numbers, and the logistics.

Outage answers are graded on instruments

Nobody is scoring your stamina at three in the morning. What carries weight is which signal told you something was wrong, what you measured before touching anything, what you rolled back versus what you fixed forward, and why you picked one. 'We restarted it and it went away' is a story about not knowing.

"How do you balance the need for scientific rigor with the pressure to ship features quickly?"

Medium · behavioural and collaboration

Approach

1. State the situation in two sentences and spend the rest on the reasoning.
2. Close with what you would do differently, concretely.
3. Pick a story where you made the decision, not one where you watched it.

Follow-up

• What would you do differently if you ran that again?
• How did you know your change caused the improvement?

Narrate an outage you owned from page to postmortem

Hard · incident response · blast radius · postmortems

Pick an incident you personally drove, ideally one where writes were affected rather than reads. In six to eight minutes: state the symptom as it first appeared on a dashboard, the blast radius you established before you knew the cause, the mitigation you applied and when, the mechanism you eventually proved, and the follow-up that would prevent a repeat. Bring numbers: error rate, tenants affected, minutes to mitigate, minutes to resolve. If you cannot name what you measured, choose a different incident.

Approach

1. Open on the signal rather than the cause: which metric at which percentile moved, on which service, at what time, so the listener follows the same evidence you had rather than a conclusion you already reached.
2. Separate mitigation from diagnosis out loud. State what you did to stop the bleeding (flag off, shed traffic, drain a lease, roll back a deploy) and say plainly that you did it before the mechanism was known, because those are two jobs with different deadlines.
3. Establish blast radius in countable terms: how many tenants, how many writes, and crucially whether the effect was loss or only delay. An append-only revision table or a pending outbox row means the change survived and the projection was merely behind, which is a repair rather than a data-loss incident.
4. Prove the mechanism instead of asserting it. Name the trace span that grew, the plan that flipped to a sequential scan, the lease that expired, plus one alternative you ruled out and the signal that stayed flat while you ruled it out.
5. Close on the durable fix and its cost, distinguishing what landed that week from what needed an expand-and-contract migration across several deploys, and say which of the two you actually finished.

Follow-up

• What would you do differently in the first five minutes, given the same dashboard and no more information?
• Which follow-up action did you deliberately not take, and why was dropping it the right call?
• How did you convince yourself the mitigation was safe to apply while the cause was still unknown?

Argue against a design, lose, and commit anyway

Medium · disagreement · service boundaries · decision records

Describe a design you argued against and lost. State the failure you predicted as a named mechanism, not a feeling about complexity: two services that would need one transaction, a projection with no rebuild path, a write path with no idempotency key. Say what evidence you brought, what the decision maker weighed instead, and what you did after the decision was made: what you instrumented, what you wrote down, and whether the prediction came true. Five minutes.

Approach

1. State the prediction in falsifiable form up front: the mechanism, the condition that triggers it, and the observable outcome. A prediction that cannot be checked also cannot be credited to you later.
2. Show the evidence you had at the time and label each piece honestly as measured, analogous, or intuition. Keeping the intuition is fine; disguising it as data is the thing that erodes your standing in the next argument.
3. Represent the opposing case at full strength, including the constraint you did not control: a fixed date, a team boundary, or the fact that the decision was cheap to reverse and yours was not.
4. Make disagree-and-commit concrete. Name the artefact you left behind so the prediction could be settled without you: the alert and its threshold, the counter on the dashboard, the decision note that recorded the trade-off and the condition that would revisit it.
5. Report the outcome without editing it. If the design held and your predicted mechanism never fired, say so and say what you had mis-weighted, which is more persuasive than a vindication story.

Follow-up

• What threshold on that alert would have proved you right, and did anyone ever look at it?
• If the same proposal arrived tomorrow with the same deadline, would you argue it the same way?
• How did you behave toward the design once it shipped and started failing in a different way than you predicted?